DistillNeRF: Perceiving 3D Scenes from Single-Glance Images by Distilling Neural Fields and Foundation Model Features

• URL: http://arxiv.org/abs/2406.12095v1
• Date: Mon, 17 Jun 2024 21:15:13 GMT
• Title: DistillNeRF: Perceiving 3D Scenes from Single-Glance Images by Distilling Neural Fields and Foundation Model Features
• Authors: Letian Wang, Seung Wook Kim, Jiawei Yang, Cunjun Yu, Boris Ivanovic, Steven L. Waslander, Yue Wang, Sanja Fidler, Marco Pavone, Peter Karkus,
• Abstract summary: DistillNeRF is a self-supervised learning framework for understanding 3D environments in autonomous driving. It predicts a rich neural scene representation from sparse, single-frame multi-view camera inputs. It is trained self-supervised with differentiable rendering to reconstruct RGB, depth, or feature images.
• Score: 65.8738034806085
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We propose DistillNeRF, a self-supervised learning framework addressing the challenge of understanding 3D environments from limited 2D observations in autonomous driving. Our method is a generalizable feedforward model that predicts a rich neural scene representation from sparse, single-frame multi-view camera inputs, and is trained self-supervised with differentiable rendering to reconstruct RGB, depth, or feature images. Our first insight is to exploit per-scene optimized Neural Radiance Fields (NeRFs) by generating dense depth and virtual camera targets for training, thereby helping our model to learn 3D geometry from sparse non-overlapping image inputs. Second, to learn a semantically rich 3D representation, we propose distilling features from pre-trained 2D foundation models, such as CLIP or DINOv2, thereby enabling various downstream tasks without the need for costly 3D human annotations. To leverage these two insights, we introduce a novel model architecture with a two-stage lift-splat-shoot encoder and a parameterized sparse hierarchical voxel representation. Experimental results on the NuScenes dataset demonstrate that DistillNeRF significantly outperforms existing comparable self-supervised methods for scene reconstruction, novel view synthesis, and depth estimation; and it allows for competitive zero-shot 3D semantic occupancy prediction, as well as open-world scene understanding through distilled foundation model features. Demos and code will be available at https://distillnerf.github.io/.

Related papers

• Denoising Diffusion via Image-Based Rendering [54.20828696348574]
  We introduce the first diffusion model able to perform fast, detailed reconstruction and generation of real-world 3D scenes. First, we introduce a new neural scene representation, IB-planes, that can efficiently and accurately represent large 3D scenes. Second, we propose a denoising-diffusion framework to learn a prior over this novel 3D scene representation, using only 2D images.
  arXiv  Detail & Related papers  (2024-02-05T19:00:45Z)
• PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm [114.47216525866435]
  We introduce a novel universal 3D pre-training framework designed to facilitate the acquisition of efficient 3D representation. For the first time, PonderV2 achieves state-of-the-art performance on 11 indoor and outdoor benchmarks, implying its effectiveness.
  arXiv  Detail & Related papers  (2023-10-12T17:59:57Z)
• Neural Feature Fusion Fields: 3D Distillation of Self-Supervised 2D Image Representations [92.88108411154255]
  We present a method that improves dense 2D image feature extractors when the latter are applied to the analysis of multiple images reconstructible as a 3D scene. We show that our method not only enables semantic understanding in the context of scene-specific neural fields without the use of manual labels, but also consistently improves over the self-supervised 2D baselines.
  arXiv  Detail & Related papers  (2022-09-07T23:24:09Z)
• Vision Transformer for NeRF-Based View Synthesis from a Single Input Image [49.956005709863355]
  We propose to leverage both the global and local features to form an expressive 3D representation. To synthesize a novel view, we train a multilayer perceptron (MLP) network conditioned on the learned 3D representation to perform volume rendering. Our method can render novel views from only a single input image and generalize across multiple object categories using a single model.
  arXiv  Detail & Related papers  (2022-07-12T17:52:04Z)
• Learning Multi-Object Dynamics with Compositional Neural Radiance Fields [63.424469458529906]
  We present a method to learn compositional predictive models from image observations based on implicit object encoders, Neural Radiance Fields (NeRFs), and graph neural networks. NeRFs have become a popular choice for representing scenes due to their strong 3D prior. For planning, we utilize RRTs in the learned latent space, where we can exploit our model and the implicit object encoder to make sampling the latent space informative and more efficient.
  arXiv  Detail & Related papers  (2022-02-24T01:31:29Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.